This Checkpoint is under construction!
Please read this post 😌
Note about v9.6:
Facing problems in results, probably I'm training a lot of concepts at the same time.
And to make matters worse, GPU prices have increased ridiculously in a short time. I had started training on an RTX 5090 paying $0.22/h and now it's over $0.50/h. The RTX 4090 is at least $0.28/h.
I appreciate the people who are sending credits on Vast.ai, but with these current GPU prices, I'll have to wait a while before continue... And if the prices don't go down, then I'll continue. Please don't forget to contribute 💜
All sample images are zero-shot generated
I will always try to avoid prompts that are similar to those used in the dataset. This way, I can show you the progression and quality of learning, making it clear that the model is learning the concepts and not just memorizing the images.
Even though no faces were blurred or removed from the images in dataset, you'll still see that it will always generate random faces (you won't see Mia Khalifa or Kendra Lust in the generated images 😌)
The images generated by this model are not at an acceptable level, so don't expect extraordinary results (yet).
❌ = Not available yet
🔴 = Worst
🟠 = Not good
🟡 = Almost there
🟢 = Good
🟣 = Perfect
Training now 🚀
Sex normalization
Penetration control
Fingering control
Refine Base NSFW
Available in beta v9.5 (preview)
Concepts to be implemented at final version:
Anus 🟡
Back view 🟡
Breasts size control (small, medium and big) 🟡
Black penis 🟢
Blowjob 🟡
Cum in mouth 🟢
Cum on face 🟢
Pussies 🟢
Doggy-style (dedicated) 🟡
Fingering 🟡
General sex positions (Missionary, Doggy, Spreading, POV, etc) 🟡
Groping ❌
Hard nipples (need more images) ❌
Naked man 🟢
Panties aside (need more images) 🟠
Playing with pussy (Rubbing, touching, etc) 🟡
POV (dedicated) 🔴
Pussy closeup 🟢
Removing clothes 🟡
See through (need more images) ❌
Penis control (flaccid, hard, small, medium, big, thin, thick, etc) 🟢
Text 🟢
Weird things (Funny and weird images, like "stepsister stucked in washing machine" 😏) 🔴
Wet clothing (need more images) ❌
Female masturbation (hands inside shorts, panties, skirts, etc) 🟠
African women (need more images) 🟢
Normalization (simple prompts for complex things) 🟠
Concepts to be incorporated soon
Sex with panties aside (dedicated)
Big breasts (to improve breasts-size control)
Sexy clothing
E-girls
Cosplay
Amateur Selfie
Wide hips (with narrow waist)
Chubby women
Fat women
Interracial sex (dedicated)
Femboys and Tomboys (dedicated)
Public (dedicated)
MILF
Lesbian (dedicated)
Teasing
Gangbang
Footjob
Cameltoe
Bathroom (dedicated)
Asian (dedicated)
Nature (dedicated)
Gay (dedicated)
Spanking/Discipline
Items marked as "dedicated" are already present in the dataset, but more images are needed for better learning.
How about helping to make the dataset?
There are many concepts and fetishes, and I don't know all of them and may not even be able to remember most of them, so you can suggest a concept to be included in the model.
You can just suggest the concept or send images/links of the concept to my HuggingFace on LoRA Requests community.
The Checkpoint Purpose:
I'm trying to bake the best SFW + NSFW Base Model.
I focused heavily on regularization so that the NSFW wouldn't end up leaking and contaminating the native SFW side of Flux. To do this, I needed to create a large regularization dataset, which significantly delays training, making it more expensive 😥
And of course, I'm not adding knowledge, I'm replacing knowledge that is less used to make room for NSFW.
My intention is to make this model incapable of producing anime-style images, so we will have "more space" for realistic NSFW content and prevent Flux from forgetting how to do other things, such as writing text and editing images (called "catastrophic forgetting").
So it's very likely that in the future we will have one Base Model just for realistic NSFW and another just for 2D-style NSFW.
Want to help me continue?
Buy PaidCoins at Aisha-AI.com and sent them to account number 2
or
Create an account on Vast.ai using my referral code (https://cloud.vast.ai/?ref_id=212422), top up your account, use the option to transfer the balance to another user and provide my ID: 212422
Total spent on this Checkpoint:
$350 USD (😭😫)
Description
Refined base for text and other stability
FAQ
Comments (18)
So far this is the best right now. the image editing is super consistent ! how many betas are left ?
@Bonticarius I have no idea how many more beta versions are to come, after all I'm adding A LOT of things and I need to run high-ranked training sessions with adjustments to keep everything working. I think maybe in a week I'll release the first version, but until it's perfect, I want to call it a beta, because the idea is to make a model as good as the original Flux, but I don't even know if I'll be able to 🙂
The Flux 2 Klein 4B trains much faster, but it's a naturally terrible model, I gave up on it.
@Bonticarius I haven't tested the image editing very much, but it's really working very well, especially considering that I haven't done ANY editing training. I'm just adding knowledge to the model for now, but maybe in the future I'll need to train how to edit the images using the face from another reference image.
Great progress. You put tremendous amound of work into it for sure. Keep it going, it will be massive in the end.
of course it will... Training now sexual interaction with penetration control. I'm tired of using LoRA sliders to set the penetration (tip, half-inserted, fully inserted)
Thanks
since you are updating checkpoint actively i would recommend charging a small fee/buzz bro. also thanks for uploading bf16 and not fp8 version. cause fp/bf16 versions easily lets users create fp8,gguf versions.
Maybe I'll do that when the checkpoint is really good. For now, these are just previews.
@AishaAI thanks for considering that but i was just saying bf16/fp16 are more versatile than fp8. fp16/bf16 versions give vivid/various paths to quantize/use,
sometimes people upload only fp8 versions which is ngl annoying which cant be converted into ggufs nor they can turned back to fp16/bf16.
so i was thanking you for uploading fp16/bf16 versions.
keep up the good work sir!
and take care of your health.
No, it doesn't easily let you create it, especially not FP8. There is a tool GGUFY but it does not create FP8 so you have to use python scripts with horrible documentation and figuring out the FP8 format is highly advanced and no clear documentation is available. Please refrain telling authors to not publish stuff that people can actually use.
I can't use this because there is no FP8 and I'm not going to bother spending an hour to dl it then convert to GGUF when my other ones are FP8.
@ferrrett33
hmm,
https://ibb.co/RGznw95k
curious if this works for you.
@opie1 Awesome, looks like they added it compared to last time I tried. Thanks a lot bro, that was very helpful.
This checkpoint is one of the best I have used, BF16 can be easily quantified as NVFP4.
Another thing is that you can use a download tool to obtain it on Huggingface. During this period, the download of Civitai has been very poor, and it no longer supports tool downloads. Moreover, there are often more than ten G checkpoints, and halfway through the download, it prompts me with "download failed", Damn!
Thanks 💜
Download only the model numbers listed here, as you will find several other versions that haven't been posted because they aren't ready yet.
@AishaAI I have a small question... How can I edit the breasts or penis in image editing ?
Enter the prompt word in image2image for whole picture? Or inpainting with area mask ?
And how should I write prompt ?
@sar090909 It depends on where you're using the template.
In Aisha-AI you use [edit:https://example.com/image.png] along with the prompt, you can use up to 3 images and you can list them in order, for example:
"Take a new photo of the woman from image 1 wearing the clothes from image 2, posing like the person in image 3".
If you are using Diffusers, there is a tutorial on the hugging face page and if you are using ComfyUI there is also an official workflow for this.
@sar090909 And one small detail... I haven't done any image editing training yet. For now, I'm focusing on teaching the model what he needs to learn. Once everything is ready, I'll actively teach him how to perform editing tasks. This means that the added features may not work correctly in the editing process, but I've done some tests and he gets it right, not with great quality, but he gets it right sometimes.
@AishaAI Yes, it works occasionally but fails most of the time. But great for use T2I. I’ll try other checkpoints or LoRAs for editing. Thanks and looking forward to the official version.